Processor and method for weight detection in a closed loop diversity mode WCDMA system

Abstract

A receiver and method for weight detection use received channel estimations and prior feedback information sent to a transmitter to estimate channel weights. The estimated channel weights may have been used by a base station to transmit diversity mode signals. In one embodiment, a trellis algorithm may be used to determine weights with no significant time-slot delay. The weights may be used to generate channel taps for use by a rake receiver in combining multipath components of received signals. In one embodiment, the receiver and method for weight detection may be suitable for use in closed loop diversity mode communication systems, including a Wideband Code Division Multiple Access (WCDMA) system.

1. A method to detect a final weight-set to process a spread spectrum channel comprising:
 determining a measurement probability for an initial weight-set from a group of predetermined weight-sets from measurements of a current time-slot; 
 determining a transition probability for the initial weight-set, the transition probability based at least on a change from a previously requested weight-set; 
 calculating a weight-set metric for the initial weight-set based at least on the measurement probability, the transition probability and a prior weight-set metric; and 
 detecting the final weight-set from the group of predetermined weight-sets based at least on the weight-set metric to process the current time-slot. 
 
     
     
       2. The method of  claim 1  further comprising using the detected final weight-set to combine multipath components of the current time-slot. 
     
     
       3. The method of  claim 1  wherein the weight-set metric is a weight-set metric for a current node of a trellis of nodes, and wherein calculating the weight-set metric comprises:
 calculating, for branches of the trellis leading to the current node, a branch metric based at least on the measurement probability and the transition probability; 
 calculating node metrics for the current node based at least on the branch metric of a branch leading to the current node and a metric of a prior node connected by the branch leading to the current node; and 
 selecting a greatest of the node metrics for the current node to correspond with the weight-set metric for the current node. 
 
     
     
       4. The method of  claim 3  wherein the group of predetermined weight-sets corresponds to nodes of the trellis. 
     
     
       5. The method of  claim 3  wherein the selecting comprises selecting a node from a plurality of nodes of the trellis having the greatest weight-set metric, each node of the plurality corresponding with one weight-set of the group of predetermined weight-sets. 
     
     
       6. The method of  claim 1  wherein prior to the determining the measurement probability, the method comprises selecting the initial weight-set from the group of predetermined weight sets, and
 wherein the measurement probability is determined for each weight-set of the group of predetermined weight-sets based at least on received amplitude and phase measurements of the current time-slot. 
 
     
     
       7. The method of  claim 1  wherein the determining the measurement probability for the initial weight-set further comprises estimating a probability for each weight-set of the group of predetermined weight-sets for the current time-slot by measuring received pilot symbols of a dedicated physical channel (DPCH) and a continuous pilot channel (CPICH) of the current time-slot. 
     
     
       8. The method of  claim 1  wherein the group of predetermined weight-sets include at least one of the weight-sets used by a base station in transmitting the current time-slot in diversity mode. 
     
     
       9. The method of  claim 1  wherein the transition probability is determined from a probability that a the initial weight-set was changed from other weight-sets of the group of predetermined weight-sets and based at least on feedback previously provided by a receiver to a transmitter. 
     
     
       10. The method of  claim 9  wherein the feedback is comprised of at least one feedback bit previously transmitted by a mobile unit for use by a base station in transmitting the current time-slot. 
     
     
       11. The method of  claim 1  further comprising determining channel taps from the final weight-set for use in combining multipath components of a channel during the current time-slot. 
     
     
       12. The method of  claim 1  wherein the initial weight-set and the final weight-set comprise weights, each weight having a phase and amplitude component. 
     
     
       13. The method of  claim 1  further comprising repeating the determining of the measurement probability and the transition probability, calculating the weight-set metric and detecting the final weight-set for a next time-slot, wherein the determining of the transition probability uses the weight-set metrics from the current time-slot as a prior time-slot in the next time-slot. 
     
     
       14. A processor comprising:
 a weight selection element to select an initial weight-set from a group of predetermined weight sets; 
 a metrics calculation element to determine a measurement probability for a the initial weight-set from measurements of a current time-slot, to determine a transition probability for the initial weight-set, the transition probability based at least on a change from a previously requested weight-set, and to calculate a weight-set metric for the initial weight-set based at least on the measurement probability, the transition probability and a prior weight-set metric; and 
 a weight detection element to select a final weight-set from the group of predetermined weight-sets based at least on the weight-set metric to process the current time-slot. 
 
     
     
       15. The processor of  claim 14  wherein the weight-set metric is a weight-set metric for a current node of a trellis of nodes,
 wherein the metrics calculation element calculates for branches of the trellis leading to the current node, a branch metric based at least on the measurement probability and the transition probabilitys, 
 wherein the metrics calculation element further calculates node metrics for the current node based at least on the branch metric of a branch leading to the current node and a metric of a prior node connected by the branch leading to the current node, and 
 wherein the weight detection element selects a greatest of the node metrics for the current node to correspond with the weight-set metric for the current node. 
 
     
     
       16. The processor of  claim 14  wherein the metrics calculation element determines the measurement probability for each weight-set of the group of predetermined weight-sets based at least on received amplitude and phase measurements of the current time-slot. 
     
     
       17. The processor of  claim 14  wherein the metrics calculation element determines the measurement probability by estimating a probability for each weight-set of the group of predetermined weight-sets for the current time-slot by measuring received pilot symbols of a dedicated physical channel (DPCH) and a continuous pilot channel (CPICH) of the current time-slot. 
     
     
       18. The processor of  claim 14  wherein the metrics calculation element determines the transition probability from a probability that the initial weight-set was changed from other weight-sets of the group of predetermined weight-sets and based at least on feedback previously provided by a receiver to a transmitter. 
     
     
       19. A machine readable medium having program instructions stored thereon for processing spread spectrum channels when executed within a digital processing device by:
 selecting an initial weight-set from a group of predetermined weight sets; 
 determining a measurement probability for the initial weight-set from measurements of a current time-slot; 
 determining a transition probability for the initial weight-set, the transition probability based at least on a change from a previously requested weight-set; 
 calculating a weight-set metric for the initial weight-set based at least on the measurement probability, the transition probability, and a prior weight-set metric; and 
 detecting a final weight-set from the group of predetermined weight-sets based at least on the weight-set metric to process the current time-slot. 
 
     
     
       20. The machine readable medium of  claim 19  wherein the weight-set metric is a weight-set metric for a current node of a trellis of nodes, and wherein calculating the weight-set metric comprises:
 calculating, for branches of the trellis leading to the current node, a branch metric based at least on the measurement probability and the transition probability; 
 calculating node metrics for the current node based at least on the branch metric of a branch leading to the current node and a metric of a prior node connected by the branch leading to the current node; and 
 selecting a greatest of the node metrics for the current node to correspond with the weight-set metric for the current node. 
 
     
     
       21. The machine readable medium of  claim 19  wherein determining the measurement probability includes determining the measurement probability for each weight-set of the group of predetermined weight-sets based at least on received amplitude and phase measurements of the current time-slot. 
     
     
       22. The machine readable medium of  claim 19  wherein determining the measurement probability for the weight-set further comprises estimating a probability for each weight-set of the group of predetermined weight-sets for the current time-slot by measuring received pilot symbols of a dedicated physical channel (DPCH) and a continuous pilot channel (CPICH) of the current time slot. 
     
     
       23. (Original) The machine readable medium of  claim 19  wherein the transition probability is determined from a probability that a weight-set was changed from other weight-sets of the group of predetermined weight-sets and based at least on feedback previously provided by a receiver to a transmitter. 
     
     
       24. A code division multiple access (CDMA) receiver comprising:
 a dedicated channel measurement element to measure characteristics of current time slots of a GDMA channel; 
 a weight selection element to select an initial weight-set from a group of predetermined weight sets; 
 a metrics calculation element to calculate metrics for the initial weight-sets estimated to have been used in transmitting each of the current time slots, the metrics calculated from the measured characteristics of the current time slots; 
 a weight detection element to select a final weight-set from the group of predetermined weight-sets based at least on the metrics for the initial weight-sets; and 
 a channel tap calculator to generate channel taps from the selected final weight-set. 
 
     
     
       25. The CDMA receiver of  claim 24  wherein the metrics calculation element calculates metrics for the initial weight-sets based at least on a measurement probability for the initial weight-set from the measured characteristics of one of the current time-slots, and
 wherein the metrics calculation element calculates a transition probability for the initial weight-set, based at least on a change from a previously requested weight-set. 
 
     
     
       26. The CDMA receiver of  claim 25  wherein each of the metrics for weight-sets is a weight-set metric for a current node of a trellis of nodes, and wherein the metrics calculation element calculates the metrics for weight-sets by calculating, for branches of the trellis leading to the current node, a branch metric based at least on the measurement and transition probabilities, and calculating node metrics for the current node based at least on the branch metric of a branch leading to the current node and a metric of a prior node connected by the branch leading to the current node, and
 wherein the weight detection element selects a greatest of the node metrics for the current node to correspond with the weight-set metric for the current node. 
 
     
     
       27. The CDMA receiver of  claim 24  further comprising:
 a pilot channel measurement element to measure the CDMA channel; and 
 a weight selection element to select a channel weight-set from a pilot channel measurement, wherein the weight selection element provides feedback to a transmitter based at least on the selected channel weight-set for use in subsequent transmissions to the CDMA receiver by the transmitter. 
 
     
     
       28. The CDMA receiver of  claim 24  further comprising:
 despreaders to despread received spread spectrum signals with spreading codes; and 
 a rake receiver to weigh and combine multipath components of the received spread spectrum signals using the channel taps provided by the channel tap calculator. 
 
     
     
       29. The CDMA receiver of  claim 24  wherein the dedicated channel measurement element, the metrics calculation element, the weight detection element, and the channel tap calculator are functional elements of a processor.